Decision_Tree_Case_Study.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": [],
      "collapsed_sections": [
        "1ohzpK4-uhJS",
        "rXvWujrUum9N"
      ],
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/alwinjmjak/ScalerProblems/blob/main/Decision_Tree_Case_Study.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "# Analyzing the Steel Plant Faults with Decision Trees\n",
        "\n",
        "## Context\n",
        "\n",
        "Welcome to the **Scaler Industrial** Development data analysis team! Our current project aims to improve the quality control process in steel manufacturing. Your expertise is requested to analyze the Steel Plant Faults dataset, utilizing Decision Trees to classify different types of steel plate faults. This task will help in automating the identification of defects and ensuring high standards in steel production.\n",
        "\n",
        "## Dataset Description\n",
        "\n",
        "The dataset you will analyze includes a variety of attributes that describe defects found in steel plates. These defects are classified into seven distinct categories, each representing a specific type of fault:\n",
        "\n",
        "### Steel Plate Fault Types:\n",
        "1. **Pastry:** Small patches or irregularities on the surface of the steel plate.\n",
        "2. **Z_Scratch:** Narrow, parallel scratches or marks in the direction of rolling.\n",
        "3. **K_Scratch:** Scratches similar to Z-scratches, but perpendicular to the rolling direction.\n",
        "4. **Stains:** Discolored or contaminated areas on the steel surface.\n",
        "5. **Dirtiness:** Presence of dirt or particulate matter on the steel surface.\n",
        "6. **Bumps:** Raised or protruding areas on the steel surface.\n",
        "7. **Other_Faults:** A broad category for faults not specifically mentioned in the other types.\n"
      ],
      "metadata": {
        "id": "Ow6YbSrgY5sy"
      }
    },
    {
      "cell_type": "code",
      "execution_count": 1,
      "metadata": {
        "id": "PM-aeWSQYi6Y",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "dc0ee1c1-1050-4921-859f-d790768b602f"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "--2024-08-09 05:15:12--  https://d2beiqkhq929f0.cloudfront.net/public_assets/assets/000/071/024/original/SteelPlant.csv\n",
            "Resolving d2beiqkhq929f0.cloudfront.net (d2beiqkhq929f0.cloudfront.net)... 18.64.229.91, 18.64.229.135, 18.64.229.172, ...\n",
            "Connecting to d2beiqkhq929f0.cloudfront.net (d2beiqkhq929f0.cloudfront.net)|18.64.229.91|:443... connected.\n",
            "HTTP request sent, awaiting response... 200 OK\n",
            "Length: 3133750 (3.0M) [text/plain]\n",
            "Saving to: ‘SteelPlant.csv’\n",
            "\n",
            "SteelPlant.csv      100%[===================>]   2.99M  --.-KB/s    in 0.08s   \n",
            "\n",
            "2024-08-09 05:15:12 (35.9 MB/s) - ‘SteelPlant.csv’ saved [3133750/3133750]\n",
            "\n"
          ]
        }
      ],
      "source": [
        "# Download the dataset\n",
        "!wget https://d2beiqkhq929f0.cloudfront.net/public_assets/assets/000/071/024/original/SteelPlant.csv"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "import warnings\n",
        "# Ignore all warnings\n",
        "warnings.filterwarnings(\"ignore\")"
      ],
      "metadata": {
        "id": "7Vxn4MtKotMZ"
      },
      "execution_count": 2,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "import pandas as pd\n",
        "\n",
        "df = pd.read_csv(\"SteelPlant.csv\")\n",
        "\n",
        "df.head()"
      ],
      "metadata": {
        "id": "xEVbaZtuY30E",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 256
        },
        "outputId": "13a124d1-ebea-413a-e188-da7165ecc18a"
      },
      "execution_count": 3,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "   id  X_Minimum  X_Maximum  Y_Minimum  Y_Maximum  Pixels_Areas  X_Perimeter  \\\n",
              "0   0        584        590     909972     909977            16            8   \n",
              "1   1        808        816     728350     728372           433           20   \n",
              "2   2         39        192    2212076    2212144         11388          705   \n",
              "3   3        781        789    3353146    3353173           210           16   \n",
              "4   4       1540       1560     618457     618502           521           72   \n",
              "\n",
              "   Y_Perimeter  Sum_of_Luminosity  Minimum_of_Luminosity  ...  \\\n",
              "0            5               2274                    113  ...   \n",
              "1           54              44478                     70  ...   \n",
              "2          420            1311391                     29  ...   \n",
              "3           29               3202                    114  ...   \n",
              "4           67              48231                     82  ...   \n",
              "\n",
              "   Orientation_Index  Luminosity_Index  SigmoidOfAreas  Pastry  Z_Scratch  \\\n",
              "0            -0.5000           -0.0104          0.1417       0          0   \n",
              "1             0.7419           -0.2997          0.9491       0          0   \n",
              "2            -0.0105           -0.0944          1.0000       0          0   \n",
              "3             0.6667           -0.0402          0.4025       0          0   \n",
              "4             0.9158           -0.2455          0.9998       0          0   \n",
              "\n",
              "   K_Scatch  Stains  Dirtiness  Bumps  Other_Faults  \n",
              "0         0       1          0      0             0  \n",
              "1         0       0          0      0             1  \n",
              "2         1       0          0      0             0  \n",
              "3         1       0          0      0             0  \n",
              "4         0       0          0      0             1  \n",
              "\n",
              "[5 rows x 35 columns]"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-1ca1e0d5-d194-4edd-b390-91d65233422e\" class=\"colab-df-container\">\n",
              "    <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>id</th>\n",
              "      <th>X_Minimum</th>\n",
              "      <th>X_Maximum</th>\n",
              "      <th>Y_Minimum</th>\n",
              "      <th>Y_Maximum</th>\n",
              "      <th>Pixels_Areas</th>\n",
              "      <th>X_Perimeter</th>\n",
              "      <th>Y_Perimeter</th>\n",
              "      <th>Sum_of_Luminosity</th>\n",
              "      <th>Minimum_of_Luminosity</th>\n",
              "      <th>...</th>\n",
              "      <th>Orientation_Index</th>\n",
              "      <th>Luminosity_Index</th>\n",
              "      <th>SigmoidOfAreas</th>\n",
              "      <th>Pastry</th>\n",
              "      <th>Z_Scratch</th>\n",
              "      <th>K_Scatch</th>\n",
              "      <th>Stains</th>\n",
              "      <th>Dirtiness</th>\n",
              "      <th>Bumps</th>\n",
              "      <th>Other_Faults</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>0</td>\n",
              "      <td>584</td>\n",
              "      <td>590</td>\n",
              "      <td>909972</td>\n",
              "      <td>909977</td>\n",
              "      <td>16</td>\n",
              "      <td>8</td>\n",
              "      <td>5</td>\n",
              "      <td>2274</td>\n",
              "      <td>113</td>\n",
              "      <td>...</td>\n",
              "      <td>-0.5000</td>\n",
              "      <td>-0.0104</td>\n",
              "      <td>0.1417</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>1</td>\n",
              "      <td>808</td>\n",
              "      <td>816</td>\n",
              "      <td>728350</td>\n",
              "      <td>728372</td>\n",
              "      <td>433</td>\n",
              "      <td>20</td>\n",
              "      <td>54</td>\n",
              "      <td>44478</td>\n",
              "      <td>70</td>\n",
              "      <td>...</td>\n",
              "      <td>0.7419</td>\n",
              "      <td>-0.2997</td>\n",
              "      <td>0.9491</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>2</td>\n",
              "      <td>39</td>\n",
              "      <td>192</td>\n",
              "      <td>2212076</td>\n",
              "      <td>2212144</td>\n",
              "      <td>11388</td>\n",
              "      <td>705</td>\n",
              "      <td>420</td>\n",
              "      <td>1311391</td>\n",
              "      <td>29</td>\n",
              "      <td>...</td>\n",
              "      <td>-0.0105</td>\n",
              "      <td>-0.0944</td>\n",
              "      <td>1.0000</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>3</td>\n",
              "      <td>781</td>\n",
              "      <td>789</td>\n",
              "      <td>3353146</td>\n",
              "      <td>3353173</td>\n",
              "      <td>210</td>\n",
              "      <td>16</td>\n",
              "      <td>29</td>\n",
              "      <td>3202</td>\n",
              "      <td>114</td>\n",
              "      <td>...</td>\n",
              "      <td>0.6667</td>\n",
              "      <td>-0.0402</td>\n",
              "      <td>0.4025</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>4</td>\n",
              "      <td>1540</td>\n",
              "      <td>1560</td>\n",
              "      <td>618457</td>\n",
              "      <td>618502</td>\n",
              "      <td>521</td>\n",
              "      <td>72</td>\n",
              "      <td>67</td>\n",
              "      <td>48231</td>\n",
              "      <td>82</td>\n",
              "      <td>...</td>\n",
              "      <td>0.9158</td>\n",
              "      <td>-0.2455</td>\n",
              "      <td>0.9998</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>5 rows × 35 columns</p>\n",
              "</div>\n",
              "    <div class=\"colab-df-buttons\">\n",
              "\n",
              "  <div class=\"colab-df-container\">\n",
              "    <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-1ca1e0d5-d194-4edd-b390-91d65233422e')\"\n",
              "            title=\"Convert this dataframe to an interactive table.\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\" viewBox=\"0 -960 960 960\">\n",
              "    <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n",
              "  </svg>\n",
              "    </button>\n",
              "\n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    .colab-df-buttons div {\n",
              "      margin-bottom: 4px;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "    <script>\n",
              "      const buttonEl =\n",
              "        document.querySelector('#df-1ca1e0d5-d194-4edd-b390-91d65233422e button.colab-df-convert');\n",
              "      buttonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "      async function convertToInteractive(key) {\n",
              "        const element = document.querySelector('#df-1ca1e0d5-d194-4edd-b390-91d65233422e');\n",
              "        const dataTable =\n",
              "          await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                    [key], {});\n",
              "        if (!dataTable) return;\n",
              "\n",
              "        const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "          '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "          + ' to learn more about interactive tables.';\n",
              "        element.innerHTML = '';\n",
              "        dataTable['output_type'] = 'display_data';\n",
              "        await google.colab.output.renderOutput(dataTable, element);\n",
              "        const docLink = document.createElement('div');\n",
              "        docLink.innerHTML = docLinkHtml;\n",
              "        element.appendChild(docLink);\n",
              "      }\n",
              "    </script>\n",
              "  </div>\n",
              "\n",
              "\n",
              "<div id=\"df-bbbca5b4-89f1-4e75-b1ba-702f04fc9511\">\n",
              "  <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-bbbca5b4-89f1-4e75-b1ba-702f04fc9511')\"\n",
              "            title=\"Suggest charts\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "<svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "     width=\"24px\">\n",
              "    <g>\n",
              "        <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n",
              "    </g>\n",
              "</svg>\n",
              "  </button>\n",
              "\n",
              "<style>\n",
              "  .colab-df-quickchart {\n",
              "      --bg-color: #E8F0FE;\n",
              "      --fill-color: #1967D2;\n",
              "      --hover-bg-color: #E2EBFA;\n",
              "      --hover-fill-color: #174EA6;\n",
              "      --disabled-fill-color: #AAA;\n",
              "      --disabled-bg-color: #DDD;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart {\n",
              "      --bg-color: #3B4455;\n",
              "      --fill-color: #D2E3FC;\n",
              "      --hover-bg-color: #434B5C;\n",
              "      --hover-fill-color: #FFFFFF;\n",
              "      --disabled-bg-color: #3B4455;\n",
              "      --disabled-fill-color: #666;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart {\n",
              "    background-color: var(--bg-color);\n",
              "    border: none;\n",
              "    border-radius: 50%;\n",
              "    cursor: pointer;\n",
              "    display: none;\n",
              "    fill: var(--fill-color);\n",
              "    height: 32px;\n",
              "    padding: 0;\n",
              "    width: 32px;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart:hover {\n",
              "    background-color: var(--hover-bg-color);\n",
              "    box-shadow: 0 1px 2px rgba(60, 64, 67, 0.3), 0 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "    fill: var(--button-hover-fill-color);\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart-complete:disabled,\n",
              "  .colab-df-quickchart-complete:disabled:hover {\n",
              "    background-color: var(--disabled-bg-color);\n",
              "    fill: var(--disabled-fill-color);\n",
              "    box-shadow: none;\n",
              "  }\n",
              "\n",
              "  .colab-df-spinner {\n",
              "    border: 2px solid var(--fill-color);\n",
              "    border-color: transparent;\n",
              "    border-bottom-color: var(--fill-color);\n",
              "    animation:\n",
              "      spin 1s steps(1) infinite;\n",
              "  }\n",
              "\n",
              "  @keyframes spin {\n",
              "    0% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "      border-left-color: var(--fill-color);\n",
              "    }\n",
              "    20% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    30% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    40% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    60% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    80% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "    90% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "  }\n",
              "</style>\n",
              "\n",
              "  <script>\n",
              "    async function quickchart(key) {\n",
              "      const quickchartButtonEl =\n",
              "        document.querySelector('#' + key + ' button');\n",
              "      quickchartButtonEl.disabled = true;  // To prevent multiple clicks.\n",
              "      quickchartButtonEl.classList.add('colab-df-spinner');\n",
              "      try {\n",
              "        const charts = await google.colab.kernel.invokeFunction(\n",
              "            'suggestCharts', [key], {});\n",
              "      } catch (error) {\n",
              "        console.error('Error during call to suggestCharts:', error);\n",
              "      }\n",
              "      quickchartButtonEl.classList.remove('colab-df-spinner');\n",
              "      quickchartButtonEl.classList.add('colab-df-quickchart-complete');\n",
              "    }\n",
              "    (() => {\n",
              "      let quickchartButtonEl =\n",
              "        document.querySelector('#df-bbbca5b4-89f1-4e75-b1ba-702f04fc9511 button');\n",
              "      quickchartButtonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "    })();\n",
              "  </script>\n",
              "</div>\n",
              "\n",
              "    </div>\n",
              "  </div>\n"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "dataframe",
              "variable_name": "df"
            }
          },
          "metadata": {},
          "execution_count": 3
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "# Assesment for DT Lecture 1"
      ],
      "metadata": {
        "id": "1ohzpK4-uhJS"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "\n",
        "\n",
        "---\n",
        "\n",
        "## Decision Tree parametric algo\n",
        "\n",
        "#### Task:  \n",
        "Before we can utilize DT to classify steel plate faults, let's revisit the fundamental characteristics of the algorithm.\n",
        "\n",
        "#### Question:  \n",
        "Is the Decision Tree a parametric algorithm?\n",
        "\n",
        "#### Options:\n",
        "\n",
        "A] Yes\n",
        "\n",
        "B] No"
      ],
      "metadata": {
        "id": "5a2krDSxbc09"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "\n",
        "\n",
        "---\n",
        "\n",
        "## Root Entropy for Decision Tree\n",
        "\n",
        "#### Context:\n",
        "In a Decision Tree classifier, entropy is a crucial measure used to determine the impurity of a node. In the context of the Steel Plant Faults dataset, understanding how to calculate the root entropy will provide insights into the initial uncertainty before any splits.\n",
        "\n",
        "#### Task:\n",
        "Write Python code to transform one-hot encoded fault types into a single 'Target' column and calculate the root entropy. This process involves aggregating multiple binary fault type columns into a single categorical column and then applying the entropy formula.\n",
        "\n",
        "#### Instructions:\n",
        "1. **Create the 'Target' Column:** Combine the one-hot encoded fault type columns ('Pastry', 'Z_Scratch', 'K_Scratch', 'Stains', 'Dirtiness', 'Bumps', 'Other_Faults') into a single 'Target' column that contains the fault type with the maximum value for each row.\n",
        "2. **Entropy Calculation Function:** Define a function `entropy_calc` that calculates the entropy for the 'Target' column, utilizing the probabilities of each fault type.\n",
        "3. **Compute Root Entropy:** Apply the `entropy_calc` function to the 'Target' column to find the root entropy and print the result.\n",
        "\n",
        "#### Question:\n",
        "For this task, how do you transform the one-hot encoded fault types into a single 'Target' column and calculate the root entropy?\n",
        "\n"
      ],
      "metadata": {
        "id": "gSVPyLXhdo-w"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "import numpy as np\n",
        "\n",
        "fault_types = ['Pastry', 'Z_Scratch', 'K_Scatch', 'Stains', 'Dirtiness', 'Bumps', 'Other_Faults']\n",
        "df['Target'] = df[fault_types].idxmax(axis=1)\n",
        "\n",
        "# TODO: Function to calculate entropy\n",
        "def entropy_calc(y):\n",
        "    print(y)\n",
        "    probabilities = y.value_counts(normalize=True).to_numpy() # Hint: Pandas function to count the frequency for each category.\n",
        "    print(probabilities)\n",
        "    entropies = -probabilities * np.log2(probabilities)  # Using log base 2 for entropy calculation\n",
        "    return round(np.sum(entropies), 2)\n",
        "\n",
        "# Calculate root entropy\n",
        "root_entropy = entropy_calc(df['Target'])\n",
        "print(f\"The root entropy is: {root_entropy}\")"
      ],
      "metadata": {
        "id": "wDGrY8MndokY",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "638b3978-7760-40f5-b3eb-b85fe77b38cb"
      },
      "execution_count": 4,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "0              Stains\n",
            "1        Other_Faults\n",
            "2            K_Scatch\n",
            "3            K_Scatch\n",
            "4        Other_Faults\n",
            "             ...     \n",
            "19214          Stains\n",
            "19215    Other_Faults\n",
            "19216          Pastry\n",
            "19217    Other_Faults\n",
            "19218           Bumps\n",
            "Name: Target, Length: 19219, dtype: object\n",
            "[0.34028826 0.24772361 0.17852126 0.11884073 0.05983662 0.02955409\n",
            " 0.02523544]\n",
            "The root entropy is: 2.36\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "\n",
        "\n",
        "---\n",
        "\n",
        "## Preprocessing for DT\n",
        "\n",
        "\n",
        "#### Context:\n",
        "In the preparation phase for applying a Decision Tree (DT) classifier to the Steel Plant Faults dataset, various preprocessing steps are considered. It's crucial to identify which of these steps is unnecessary for DT, given its unique properties and how it handles data.\n",
        "\n",
        "#### Question:\n",
        "Which preprocessing step can be generally avoided when implementing a Decision Tree for the Steel Plant Faults dataset, where features are on different scales?\n",
        "\n",
        "#### Options:\n",
        "A) Feature Selection\n",
        "\n",
        "B) Feature Scaling\n",
        "\n",
        "C) SMOTE\n",
        "\n",
        "D) Encoding\n",
        "\n"
      ],
      "metadata": {
        "id": "J5rnU4GSmBku"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "\n",
        "\n",
        "---\n",
        "## Understanding Decision Tree Characteristics\n",
        "\n",
        "#### Context:\n",
        "In building a Decision Tree (DT) for classification, various concepts and methods play a crucial role in the tree's construction and decision-making process. Understanding these aspects is essential for effectively applying DT in practical scenarios.\n",
        "\n",
        "#### Question:\n",
        "Which of the following statements are correct regarding the construction and characteristics of a Decision Tree?\n",
        "\n",
        "#### Statements to Evaluate:\n",
        "S1: Fill the missing attribute value with the Central Tendency (mean, median, mode) of the attribute.\n",
        "\n",
        "S2: Decision Trees can handle both numerical and categorical data.\n",
        "\n",
        "S3: The root Node is the top-most node of the Tree from where the Tree starts.\n",
        "\n",
        "S4: Information gain is the difference between the entropy of the parent and the weighted average entropy of the children.\n",
        "\n",
        "#### Options:\n",
        "A) S1, S2, S4\n",
        "\n",
        "B) S2, S3, S4\n",
        "\n",
        "C) S1, S2, S3\n",
        "\n",
        "D) S1, S2, S3, S4\n",
        "\n"
      ],
      "metadata": {
        "id": "DRO_9Zq8mN-O"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "\n",
        "\n",
        "---\n",
        "\n",
        "## Assessing Dataset Balance for Decision Tree Implementation\n",
        "\n",
        "#### Context:\n",
        "Understanding the distribution of classes in a dataset is essential for machine learning tasks, particularly for classification. The balance or imbalance of the dataset can influence the performance and decision-making process of a Decision Tree (DT) classifier.\n",
        "\n",
        "#### Task:\n",
        "You are asked to write a Python script that visualizes the distribution of classes in the Steel Plant Faults dataset to assess its balance. Based on this analysis, you will determine the necessity of applying imbalance-handling techniques when implementing a DT.\n",
        "\n",
        "#### Instructions:\n",
        "1. **Visualize Class Distribution:** Create a bar plot to visualize the count of each fault type ('Pastry', 'Z_Scratch', 'K_Scratch', 'Stains', 'Dirtiness', 'Bumps', 'Other_Faults') in the dataset.\n",
        "2. **Analyze the Plot:** Examine the bar plot to assess whether the dataset is balanced or imbalanced across the different fault types.\n",
        "3. **Decision Tree Consideration:** Decide whether a DT classifier would require imbalance-handling techniques based on the class distribution.\n",
        "\n",
        "#### Question:\n",
        "Based on your analysis of the class distribution in the Steel Plant Faults dataset, is the dataset balanced, and should an imbalance-handling algorithm be applied when implementing a Decision Tree?\n",
        "\n",
        "#### Options:\n",
        "A) The dataset is balanced, and Decision Trees are not affected by data imbalance.\n",
        "\n",
        "B) The dataset is imbalanced, but Decision Trees are not significantly affected by data imbalance.\n",
        "\n",
        "C) The dataset is balanced, but Decision Trees are affected by data imbalance.\n",
        "\n",
        "D) The dataset is imbalanced, and Decision Trees are affected by data imbalance."
      ],
      "metadata": {
        "id": "LxPNGGlSn7s9"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "import seaborn as sns\n",
        "import matplotlib.pyplot as plt\n",
        "\n",
        "fault_types = ['Pastry', 'Z_Scratch', 'K_Scatch', 'Stains', 'Dirtiness', 'Bumps', 'Other_Faults']\n",
        "\n",
        "# TODO: Count number of data points for each fault type\n",
        "fault_counts = df['Target'].value_counts()\n",
        "\n",
        "# Set a custom color palette (optional)\n",
        "custom_palette = sns.color_palette(\"pastel\")\n",
        "\n",
        "# TODO: Create the bar plot to visualize the count of each fault type\n",
        "plt.figure(figsize=(8, 6))  # Adjust figure size\n",
        "sns.barplot(x=fault_types, y=fault_counts.to_numpy(), palette=custom_palette)\n",
        "\n",
        "# Customize labels and title\n",
        "plt.xlabel(\"Classes\", fontsize=12)\n",
        "plt.ylabel(\"Counts\", fontsize=12)\n",
        "plt.title(\"Distribution of Classes\", fontsize=14)\n",
        "\n",
        "# Show the plot\n",
        "plt.tight_layout()  # Ensure labels fit within the figure\n",
        "plt.show()"
      ],
      "metadata": {
        "id": "z4bs6xYJiGIK",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 607
        },
        "outputId": "d12fbbbf-5923-4079-d67f-e4cc916b3905"
      },
      "execution_count": 5,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 800x600 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "\n",
        "\n",
        "---\n",
        "\n",
        "## Find Root Node\n",
        "\n",
        "#### Context:\n",
        "In a Decision Tree, the root node is pivotal as it represents the initial decision point from which further splits are derived. The Gini value at this node offers insight into the node's purity. This exercise involves training a Decision Tree using only two features from the Steel Plant Faults dataset and determining the root node's feature and its Gini value.\n",
        "\n",
        "#### Task:\n",
        "Train a Decision Tree classifier using 'Pixels_Areas' and 'SigmoidOfAreas' as features. Set the `max_depth` to 3 and `random_state` to 42. Determine which feature is chosen as the root node and identify its Gini value.\n",
        "\n",
        "#### Instructions:\n",
        "1. **Feature Selection:** Use 'Pixels_Areas' and 'SigmoidOfAreas' as your features for the Decision Tree.\n",
        "2. **Data Splitting:** Split the dataset into training and testing sets.\n",
        "3. **Model Initialization:** Initialize a Decision Tree Classifier with `max_depth=3` and `random_state=42`.\n",
        "4. **Model Training:** Train the Decision Tree on the training data.\n",
        "5. **Root Node Identification:** Identify the feature used at the root node and find its Gini value.\n",
        "\n",
        "#### Question:\n",
        "After training the Decision Tree with the specified hyperparameters using 'Pixels_Areas' and 'SigmoidOfAreas' as features, which feature becomes the root node, and what is its Gini value?\n",
        "\n",
        "#### Options:\n",
        "A) 'Pixels_Areas' is the root node, with a Gini value of X.\n",
        "\n",
        "B) 'SigmoidOfAreas' is the root node, with a Gini value of Y.\n",
        "\n",
        "C) Both 'Pixels_Areas' and 'SigmoidOfAreas' share the root node, with Gini values of X and Y respectively.\n",
        "\n",
        "D) Neither 'Pixels_Areas' nor 'SigmoidOfAreas' becomes the root node."
      ],
      "metadata": {
        "id": "inRu78UQzFEf"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "from sklearn.tree import DecisionTreeClassifier\n",
        "from sklearn.model_selection import train_test_split\n",
        "\n",
        "# TODO: Selecting the two features for training\n",
        "X = df[['Pixels_Areas', 'SigmoidOfAreas']]\n",
        "y = df['Target']\n",
        "\n",
        "# TODO: Split the dataset into training and testing sets\n",
        "X_train, X_test, y_train, y_test = train_test_split(X, y, random_state=42)\n",
        "\n",
        "# TODO: Initialize the Decision Tree Classifier with maximum depth if 3 and random_state=42\n",
        "dt_classifier = DecisionTreeClassifier(max_depth=3, random_state=42)\n",
        "\n",
        "# TODO: Train the Decision Tree Classifier on the training data\n",
        "dt_classifier.fit(X_train, y_train)\n",
        "\n",
        "# TODO: Getting the root node feature name and its Gini value\n",
        "# Hint - use documentation to find the property to access the values.\n",
        "root_node_feature_index = dt_classifier.tree_.feature[0]\n",
        "root_node_feature_name = X.columns[root_node_feature_index]\n",
        "root_node_gini_value = dt_classifier.tree_.impurity[0]\n",
        "\n",
        "print(f\"The root node feature is: {root_node_feature_name}\")\n",
        "print(f\"The Gini value of the root node is: {root_node_gini_value:.2f}\")"
      ],
      "metadata": {
        "id": "L-lqBteVoHJk",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "19f8becc-7e5f-481c-86f6-146c579138ae"
      },
      "execution_count": 6,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "The root node feature is: Pixels_Areas\n",
            "The Gini value of the root node is: 0.77\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Gini of 5 purple and 5 red balls being split\n",
        "1 - ((.5)**2 + (.5)**2)"
      ],
      "metadata": {
        "id": "lInd_bgRuGF-",
        "outputId": "144e269f-7315-4da6-a4b0-032b4cb5379e",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "execution_count": 52,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.5"
            ]
          },
          "metadata": {},
          "execution_count": 52
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "1 - ((4/5)**2)"
      ],
      "metadata": {
        "id": "50mbGIh6uGMw",
        "outputId": "59943240-e23c-46d3-da2d-7f54c35fe963",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "execution_count": 53,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.3599999999999999"
            ]
          },
          "metadata": {},
          "execution_count": 53
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "1 - ((1/5)**2 + (5/5)**2)"
      ],
      "metadata": {
        "id": "4Djwda_0uGUO",
        "outputId": "3be98e9c-54f1-4504-c2e1-d38d059546df",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "execution_count": 54,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "-0.040000000000000036"
            ]
          },
          "metadata": {},
          "execution_count": 54
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "\n",
        "\n",
        "---\n",
        "\n",
        "# Assesment for DT Lecture 2"
      ],
      "metadata": {
        "id": "rXvWujrUum9N"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "\n",
        "\n",
        "---\n",
        "## DT- Root Split\n",
        "\n",
        "#### Context:  \n",
        "Understanding the sensitivity of a Decision Tree's structure to changes in the training data can provide insights into the model's stability and robustness. Removing a data point from the dataset may affect the tree's formation, particularly the feature and threshold selected for the root node's split.\n",
        "\n",
        "#### Task:  \n",
        "Consider a scenario where you remove one data point from the dataset used to train a Decision Tree.\n",
        "\n",
        "#### Question:  \n",
        "Which option is true regarding the splitting criteria at the root of the tree if one of the data points is removed?\n",
        "\n",
        "#### Options:  \n",
        "A) The splitting criteria at the root will be the same.  \n",
        "B) The splitting criteria at the root will be different.  \n",
        "C) The splitting criteria at the root could be the same or different.  \n",
        "D) The splitting criteria at the root can’t be related to this removal of the data point.\n",
        "\n",
        "\n"
      ],
      "metadata": {
        "id": "pzYHNsmMuWiO"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "\n",
        "\n",
        "---\n",
        "\n",
        "## Most Misclassified DT\n",
        "\n",
        "#### Context:\n",
        "Using class weights in a Decision Tree can help address class imbalance by adjusting the decision criteria toward less represented classes. This task involves applying class weights to a Decision Tree classifier and identifying the class that is most frequently misclassified.\n",
        "\n",
        "#### Task:\n",
        "Write Python code to train a Decision Tree classifier using entropy as the criterion and balanced class weights. Your goal is to determine which class is most misclassified on the test dataset.\n",
        "\n",
        "#### Instructions:\n",
        "1. **Feature and Target Selection:** Prepare your features `X` by dropping the class columns and 'Target', and set `y` as your 'Target'.\n",
        "2. **Data Splitting:** Split the dataset into training and testing sets using `train_test_split` with `random_state=42`.\n",
        "3. **Class Weights Calculation:** Use the 'balanced' option for class weights to handle class imbalance.\n",
        "4. **Initialize and Train Decision Tree:** Create a Decision Tree classifier with entropy criterion, incorporating the calculated class weights. Train this classifier on your training data.\n",
        "5. **Model Prediction and Evaluation:** Use the trained model to predict the classes on the test data. Generate a classification report to evaluate the model's performance across different classes.\n",
        "6. **Identify Most Misclassified Class:** Analyze the recall values from the classification report to identify which class is the most misclassified.\n",
        "\n",
        "#### Question:\n",
        "After applying the described Decision Tree configuration and analyzing the classification report, which class is identified as the most misclassified?\n",
        "\n",
        "#### Options:\n",
        "A) Bumps\n",
        "\n",
        "B) Other_Faults\n",
        "\n",
        "C) Dirtiness\n",
        "\n",
        "D) Pastry\n",
        "\n"
      ],
      "metadata": {
        "id": "tY_Y08WHuuTm"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "df.columns"
      ],
      "metadata": {
        "id": "ry4g8k9Q01DY",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "7e07a172-9a21-4d22-982c-46171fd0c67e"
      },
      "execution_count": 55,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Index(['id', 'X_Minimum', 'X_Maximum', 'Y_Minimum', 'Y_Maximum',\n",
              "       'Pixels_Areas', 'X_Perimeter', 'Y_Perimeter', 'Sum_of_Luminosity',\n",
              "       'Minimum_of_Luminosity', 'Maximum_of_Luminosity', 'Length_of_Conveyer',\n",
              "       'TypeOfSteel_A300', 'TypeOfSteel_A400', 'Steel_Plate_Thickness',\n",
              "       'Edges_Index', 'Empty_Index', 'Square_Index', 'Outside_X_Index',\n",
              "       'Edges_X_Index', 'Edges_Y_Index', 'Outside_Global_Index', 'LogOfAreas',\n",
              "       'Log_X_Index', 'Log_Y_Index', 'Orientation_Index', 'Luminosity_Index',\n",
              "       'SigmoidOfAreas', 'Pastry', 'Z_Scratch', 'K_Scatch', 'Stains',\n",
              "       'Dirtiness', 'Bumps', 'Other_Faults', 'Target'],\n",
              "      dtype='object')"
            ]
          },
          "metadata": {},
          "execution_count": 55
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "from sklearn.tree import DecisionTreeClassifier\n",
        "from sklearn.metrics import classification_report\n",
        "\n",
        "# Prepare our features for training\n",
        "X = df.drop(columns=['Pastry', 'Z_Scratch', 'K_Scatch', 'Stains',\n",
        "       'Dirtiness', 'Bumps', 'Other_Faults', 'Target'], axis =1)\n",
        "y = df['Target']\n",
        "\n",
        "# Split the dataset into training and testing sets\n",
        "X_train, X_test, y_train, y_test = train_test_split(X, y, random_state=42)\n",
        "\n",
        "# TODO: Initialize the Decision Tree classifier with the given hyperparameters and class weights\n",
        "# Hint: Use documentation to find how to balance the class_weights in DecisionTrees\n",
        "dt_classifier = DecisionTreeClassifier(criterion='entropy', random_state=42, class_weight='balanced')\n",
        "\n",
        "# TODO: Train the Decision Tree classifier on training data\n",
        "dt_classifier.fit(X_train, y_train)\n",
        "\n",
        "# TODO: Predict on the test data\n",
        "y_pred = dt_classifier.predict(X_test)\n",
        "\n",
        "# TODO: Generate a classification report to include the metrics for each class\n",
        "report = classification_report(y_test, y_pred, target_names=np.unique(y_train))\n",
        "print(report)"
      ],
      "metadata": {
        "id": "ORVdFsFIvrBq",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "ae1768a9-a52c-4e2e-9934-794b0457125f"
      },
      "execution_count": 7,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "              precision    recall  f1-score   support\n",
            "\n",
            "       Bumps       0.39      0.39      0.39      1165\n",
            "   Dirtiness       0.14      0.18      0.16       107\n",
            "    K_Scatch       0.83      0.83      0.83       852\n",
            "Other_Faults       0.41      0.39      0.40      1671\n",
            "      Pastry       0.22      0.21      0.22       605\n",
            "      Stains       0.62      0.65      0.63       130\n",
            "   Z_Scratch       0.40      0.44      0.42       275\n",
            "\n",
            "    accuracy                           0.45      4805\n",
            "   macro avg       0.43      0.44      0.44      4805\n",
            "weighted avg       0.45      0.45      0.45      4805\n",
            "\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "\n",
        "\n",
        "---\n",
        "\n",
        "## Splitting Data\n",
        "\n",
        "#### Context:\n",
        "In Decision Trees, achieving a node where each child contains samples exclusively from one class represents an ideal split where classes are perfectly separated. Such scenarios prompt specific decisions on how to proceed with further splits or terminate the tree building process.\n",
        "\n",
        "#### Task:\n",
        "You have a Decision Tree where the feature `X_minimum` leads to a node split such that each of its 7 children contains samples belonging exclusively to one class. Decide the next steps in the tree construction process.\n",
        "\n",
        "#### Question:\n",
        "If using the feature `X_minimum` results in each child node containing samples from exactly one class, what should you do next?\n",
        "\n",
        "#### Options:\n",
        "A) Find another feature to split the node further.\n",
        "\n",
        "B) Declare it as a leaf node since it is a pure node.\n",
        "\n",
        "C) Always terminate the recursions on all branches and return the current tree.\n",
        "\n",
        "D) Go back to the parent node and select a different feature to split the data so that y-values are not all the same at this node.\n"
      ],
      "metadata": {
        "id": "Uhga3yhe5T8K"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "\n",
        "\n",
        "---\n",
        "\n",
        "## Decision Trees Characteristics\n",
        "\n",
        "#### Context:  \n",
        "Decision Trees are a fundamental machine learning algorithm used for both classification and regression tasks. Understanding their characteristics, capabilities, and limitations is crucial for effectively applying them to solve real-world problems.\n",
        "\n",
        "#### Question:  \n",
        "Which of the following statements are true regarding the properties and behavior of Decision Trees?\n",
        "\n",
        "#### Statements to Evaluate:  \n",
        "1\\. Decision tree makes no assumptions about the data.  \n",
        "2\\. The decision tree model can learn non-linear decision boundaries.  \n",
        "3\\. Decision trees cannot explain how the target will change if a variable is changed by 1 unit (marginal effect).  \n",
        "4\\. Hyperparameter tuning is not required in decision trees.  \n",
        "5\\. In a decision tree, increasing entropy implies increasing purity.  \n",
        "6\\. In a decision tree, the entropy of a node decreases as we go down the decision tree.\n",
        "\n",
        "#### Options:  \n",
        "A) 1, 2, and 5  \n",
        "B) 3, 5, and 6  \n",
        "C) 2, 3, 4, and 5  \n",
        "D) 1, 2, 3, and 6"
      ],
      "metadata": {
        "id": "U2efMO6j8VXH"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "\n",
        "\n",
        "---\n",
        "## Optimizing Decision Tree Depth\n",
        "\n",
        "#### Context:\n",
        "Cross-validation is an essential technique for assessing the generalizability of machine learning models. In this task, you will use k-fold cross-validation to determine the optimal depth for a Decision Tree classifier that maximizes accuracy on a given dataset.\n",
        "\n",
        "#### Task:\n",
        "Write Python code to perform k-fold cross-validation on a series of Decision Tree models with varying depths. The goal is to identify the tree depth that results in the best performance on the test data.\n",
        "\n",
        "#### Instructions:\n",
        "1. **Define Depth Values:** Use the list of depths `[3, 5, 7, 11, 13, 15, 19, 20]` to evaluate which tree depth provides the best model performance.\n",
        "2. **Initialize K-Fold Cross-Validation:** Set up a `KFold` instance with 10 splits and `random_state=42` to ensure reproducibility.\n",
        "3. **Cross-Validation Loop:** For each depth in your list, initialize a Decision Tree with `class_weight='balanced'`. Perform cross-validation and record the training and validation accuracies.\n",
        "4. **Determine Best Depth:** Analyze the cross-validation results to select the depth that provides the highest validation accuracy.\n",
        "5. **Train and Evaluate the Best Model:** Using the identified best depth, train a new Decision Tree on the entire training set and evaluate its performance on the test set.\n",
        "\n",
        "#### Question:\n",
        "Based on your cross-validation results, which depth should be selected for the Decision Tree to achieve optimal performance, and what is the test accuracy for this depth?"
      ],
      "metadata": {
        "id": "sCkDaYHm9K8d"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "import numpy as np\n",
        "from sklearn.tree import DecisionTreeClassifier\n",
        "from sklearn.model_selection import KFold, cross_validate\n",
        "\n",
        "# Define the list of tree depths you want to evaluate\n",
        "depths = [3, 5, 7, 11, 13, 15, 19, 20]\n",
        "\n",
        "# TODO: Setup KFold with 10 splits and a fixed random state\n",
        "kfold = KFold(n_splits=10, shuffle=True, random_state=42)\n",
        "\n",
        "for depth in depths:\n",
        "    # TODO: Initialize Decision Tree Classifier with the current depth\n",
        "    tree_clf = DecisionTreeClassifier(max_depth=depth, random_state=42, class_weight='balanced')\n",
        "\n",
        "    # TODO: Perform cross-validation\n",
        "    cv_acc_results = cross_validate(tree_clf, X_train, y_train, cv=kfold, scoring='accuracy', return_train_score=True)\n",
        "\n",
        "    # Output the mean and standard deviation of the training and validation scores\n",
        "    print(f\"K-Fold for depth: {depth} - Accuracy Mean: Train: {cv_acc_results['train_score'].mean() * 100:.2f}%, Validation: {cv_acc_results['test_score'].mean() * 100:.2f}%\")\n",
        "    print(f\"K-Fold for depth: {depth} - Accuracy Std: Train: {cv_acc_results['train_score'].std() * 100:.2f}%, Validation: {cv_acc_results['test_score'].std() * 100:.2f}%\")\n",
        "    print('***************')\n",
        "\n",
        "# Testing the best depth on the testing data\n",
        "best_tree_clf = DecisionTreeClassifier(max_depth=5, random_state=42, class_weight='balanced')\n",
        "best_tree_clf.fit(X_train, y_train)\n",
        "print(f\"The model accuracy on testing data with best depth is {best_tree_clf.score(X_test, y_test)}\")"
      ],
      "metadata": {
        "id": "oZ8WjnwixQt-",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "6508550b-2ca1-4d83-853b-c56549bcee23"
      },
      "execution_count": 8,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "K-Fold for depth: 3 - Accuracy Mean: Train: 42.29%, Validation: 42.19%\n",
            "K-Fold for depth: 3 - Accuracy Std: Train: 0.32%, Validation: 1.25%\n",
            "***************\n",
            "K-Fold for depth: 5 - Accuracy Mean: Train: 50.11%, Validation: 49.67%\n",
            "K-Fold for depth: 5 - Accuracy Std: Train: 1.49%, Validation: 1.45%\n",
            "***************\n",
            "K-Fold for depth: 7 - Accuracy Mean: Train: 46.70%, Validation: 44.65%\n",
            "K-Fold for depth: 7 - Accuracy Std: Train: 0.83%, Validation: 1.41%\n",
            "***************\n",
            "K-Fold for depth: 11 - Accuracy Mean: Train: 57.57%, Validation: 46.70%\n",
            "K-Fold for depth: 11 - Accuracy Std: Train: 0.65%, Validation: 0.88%\n",
            "***************\n",
            "K-Fold for depth: 13 - Accuracy Mean: Train: 65.73%, Validation: 48.13%\n",
            "K-Fold for depth: 13 - Accuracy Std: Train: 0.42%, Validation: 1.43%\n",
            "***************\n",
            "K-Fold for depth: 15 - Accuracy Mean: Train: 74.58%, Validation: 48.04%\n",
            "K-Fold for depth: 15 - Accuracy Std: Train: 0.77%, Validation: 1.23%\n",
            "***************\n",
            "K-Fold for depth: 19 - Accuracy Mean: Train: 88.18%, Validation: 48.24%\n",
            "K-Fold for depth: 19 - Accuracy Std: Train: 0.93%, Validation: 1.33%\n",
            "***************\n",
            "K-Fold for depth: 20 - Accuracy Mean: Train: 90.35%, Validation: 48.10%\n",
            "K-Fold for depth: 20 - Accuracy Std: Train: 0.89%, Validation: 1.27%\n",
            "***************\n",
            "The model accuracy on testing data with best depth is 0.495525494276795\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "\n",
        "\n",
        "---\n",
        "\n",
        "## DT - Feature Importance\n",
        "\n",
        "#### Context:\n",
        "Understanding which features most significantly impact the predictions of a Decision Tree is crucial for refining models and enhancing interpretability. This task involves identifying the top three features based on their importance scores from a trained Decision Tree.\n",
        "\n",
        "#### Task:\n",
        "Write Python code to analyze the feature importance from a Decision Tree (`best_tree_clf`) and identify the top three most influential features.\n",
        "\n",
        "#### Instructions:\n",
        "1. **Retrieve Feature Importances:** Extract the importance scores of all features from the `best_tree_clf`.\n",
        "2. **Rank Features by Importance:** Sort the features by their importance in descending order.\n",
        "3. **Identify Top Three Features:** Determine the three features with the highest importance scores.\n",
        "4. **Visualize Feature Importance:** Optionally, create a bar plot to visually represent the importance of each feature, which can aid in understanding their relative contributions to the model's decisions.\n",
        "\n",
        "#### Question:\n",
        "After executing your script to analyze feature importances in the Decision Tree, which three features are identified as the most influential based on their importance scores?\n",
        "\n",
        "#### Options:\n",
        "A) `Outside_X_Index`, `Pixels_Areas`, `Length_of_Conveyer`\n",
        "\n",
        "B) `X_Minimum`, `Y_Maximum`, `Sum_of_Luminosity`\n",
        "\n",
        "C) `Maximum_of_Luminosity`, `Edges_Index`, `Steel_Plate_Thickness`\n",
        "\n",
        "D) `TypeOfSteel_A300`, `Edges_Y_Index`, `LogOfAreas`\n",
        "\n"
      ],
      "metadata": {
        "id": "FUq956RvB5Hz"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "feature_names = X_train.columns\n",
        "\n",
        "# TODO: Get feature importances of the best_tree_clf from 'previous question'\n",
        "feature_importances = best_tree_clf.feature_importances_\n",
        "# TODO: Sort the features by importance (get their indices only not the values. )\n",
        "sorted_idx = np.argsort(feature_importances)[::-1]  # This sorts indices by importance from high to low\n",
        "# Output the top 3 features\n",
        "print(\"Top 3 features:\")\n",
        "for i in range(3):\n",
        "    feature_index = sorted_idx[i]\n",
        "    print(f\"{i+1}. {feature_names[feature_index]} with an importance of {feature_importances[feature_index]:.4f}\")\n",
        "\n",
        "sorted_feature_names = [feature_names[i] for i in sorted_idx]\n",
        "sorted_importances = feature_importances[sorted_idx]\n",
        "\n",
        "# TODO: Create a horizontal bar plot for the feature importances\n",
        "plt.figure(figsize=(10, 6))\n",
        "plt.barh(range(len(sorted_importances)), sorted_importances, align='center', color='dodgerblue')\n",
        "plt.yticks(range(len(sorted_importances)), sorted_feature_names)\n",
        "plt.xlabel('Importance')\n",
        "plt.ylabel('Feature Names')\n",
        "plt.title('Feature Importance from Decision Tree')\n",
        "plt.gca().invert_yaxis()  # Invert y-axis to have the highest importance on top\n",
        "plt.show()"
      ],
      "metadata": {
        "id": "MaiJwubO-uGG",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 633
        },
        "outputId": "859a5d91-59f9-493b-8621-ce78e1f99e3d"
      },
      "execution_count": 9,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Top 3 features:\n",
            "1. Outside_X_Index with an importance of 0.3198\n",
            "2. Pixels_Areas with an importance of 0.2949\n",
            "3. Length_of_Conveyer with an importance of 0.1254\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1000x600 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "\n",
        "\n",
        "---\n",
        "\n",
        "# Assesment for Bagging\n"
      ],
      "metadata": {
        "id": "fqR-fRBMFXcW"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Role of Ensemble Methods\n",
        "\n",
        "\n",
        "#### Question:\n",
        "Choose the correct statements concerning the use of ensemble algorithms with Decision Trees in machine learning.\n",
        "\n",
        "#### Statements to Evaluate:\n",
        "S1: To improve our Decision Tree model performance, an ensemble algorithm is used.\n",
        "S2: The Decision Tree model in our dataset was overfitting, hence an ensemble algorithm was used.\n",
        "S3: Ensemble uses many independent models and aggregates their output.\n",
        "S4: The Decision Tree model in our dataset was underfitting, hence an ensemble algorithm was used.\n",
        "\n",
        "#### Options:\n",
        "A) S1, S2, S3\n",
        "\n",
        "B) S1, S3, S4\n",
        "\n",
        "C) S2, S3, S4\n",
        "\n",
        "D) All of the above\n"
      ],
      "metadata": {
        "id": "N28YS0n9bWSL"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "\n",
        "\n",
        "---\n",
        "\n",
        "## SMOTE imbalance\n",
        "\n",
        "\n",
        "#### Context:\n",
        "The Synthetic Minority Over-sampling Technique (SMOTE) is employed to handle class imbalance by generating synthetic samples for minority classes. This technique is crucial for ensuring that machine learning models do not become biased towards the majority class.\n",
        "\n",
        "#### Task:\n",
        "Using the SMOTE technique, determine the resulting sample size for each class after applying the algorithm. You are to predict the uniform size that each class will have post-SMOTE, assuming SMOTE balances all classes to the size of the originally largest class.\n",
        "\n",
        "#### Instructions:\n",
        "1. **Review Initial Class Distribution:** Start by analyzing the initial distribution of your classes. This will help you understand the extent of the imbalance you need to address.\n",
        "2. **Apply SMOTE:** Implement the SMOTE technique on your dataset to balance the class distribution. Use `random_state=42` for reproducibility.\n",
        "3. **Analyze Post-SMOTE Distribution:** After applying SMOTE, examine the new class sizes. All classes should now reflect the size of the largest original class if SMOTE has been applied correctly.\n",
        "4. **Determine Uniform Class Size:** From the balanced dataset, identify the uniform size to which all classes have been adjusted.\n",
        "\n",
        "#### Question:\n",
        "What will be the sample size for each class after applying SMOTE, given that SMOTE adjusts all classes to the size of the largest class present before balancing?\n",
        "\n",
        "#### Options:\n",
        "A) 6540\n",
        "\n",
        "B) 5530\n",
        "\n",
        "C) 7000\n",
        "\n",
        "D) 4210\n",
        "\n"
      ],
      "metadata": {
        "id": "cNJU-HkgdDfl"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "from imblearn.over_sampling import SMOTE\n",
        "\n",
        "# TODO: Count class instances before SMOTE from the variable ('y')\n",
        "print(\"\\Before SMOTE:\")\n",
        "print(y.value_counts())\n",
        "\n",
        "# TODO: Apply SMOTE\n",
        "smote = SMOTE(random_state=42)\n",
        "X_resampled, y_resampled = smote.fit_resample(X, y)\n",
        "\n",
        "# TODO: Count class instances after SMOTE on variable 'y_resampled'\n",
        "print(\"\\nAfter SMOTE:\")\n",
        "print(y_resampled.value_counts())"
      ],
      "metadata": {
        "id": "SF2wTOZLDsZC",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "ee3d6507-69f6-412d-bcf6-00ea38127320"
      },
      "execution_count": 10,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\\Before SMOTE:\n",
            "Target\n",
            "Other_Faults    6540\n",
            "Bumps           4761\n",
            "K_Scatch        3431\n",
            "Pastry          2284\n",
            "Z_Scratch       1150\n",
            "Stains           568\n",
            "Dirtiness        485\n",
            "Name: count, dtype: int64\n",
            "\n",
            "After SMOTE:\n",
            "Target\n",
            "Stains          6540\n",
            "Other_Faults    6540\n",
            "K_Scatch        6540\n",
            "Bumps           6540\n",
            "Pastry          6540\n",
            "Z_Scratch       6540\n",
            "Dirtiness       6540\n",
            "Name: count, dtype: int64\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "\n",
        "\n",
        "---\n",
        "## Understanding OOB\n",
        "\n",
        "\n",
        "#### Context:\n",
        "In ensemble methods such as Random Forests that utilize bootstrapping, Out-Of-Bag (OOB) data plays a critical role in estimating the model's performance without needing a separate validation set. Understanding what OOB data represents is key to effectively using these ensemble methods in practice.\n",
        "\n",
        "#### Task:\n",
        "Explain the concept of OOB data in the context of ensemble models that use bootstrapping techniques.\n",
        "\n",
        "#### Question:\n",
        "What does OOB data represent in the context of bootstrapping used in ensemble learning methods?\n",
        "\n",
        "#### Statements:\n",
        "S1:  OOB data is the remaining samples which are not selected by bootstrapping.\n",
        "\n",
        "S2:  OOB data is the remaining samples in the validation dataset.\n",
        "\n",
        "S3:  OOB data is the used samples which were selected by bootstrapping.\n",
        "\n",
        "S4:  OOB data is the used samples in the validation dataset.\n"
      ],
      "metadata": {
        "id": "tYGLlQuGjwjb"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "\n",
        "\n",
        "---\n",
        "\n",
        "## Comparing OOB Scores\n",
        "\n",
        "#### Context:\n",
        "The Out-Of-Bag (OOB) score in Random Forest models offers a unique method of validation by utilizing unsampled data points during the training process as a test set. This task involves comparing the OOB scores of a Random Forest classifier trained on original versus SMOTE-resampled data to evaluate how class balancing affects model performance.\n",
        "\n",
        "#### Task:\n",
        "Implement and compare two Random Forest classifiers: one trained on the original dataset and another on a dataset resampled using SMOTE. Your goal is to analyze the difference in OOB scores between these two approaches.\n",
        "\n",
        "#### Instructions:\n",
        "1. **Balance the Dataset Using SMOTE:** Apply SMOTE to the original dataset to balance class distributions, ensuring that minority classes are adequately represented.\n",
        "2. **Split Both Datasets:** Split both the original and the SMOTE-resampled datasets into training and testing sets using a 33% test size and a random state of 42.\n",
        "3. **Initialize and Train Two Random Forests:** Create two Random Forest classifiers with `oob_score=True` and `random_state=42`. Train one on the original training set and another on the resampled training set.\n",
        "4. **Retrieve and Compare OOB Scores:** After training, extract the OOB scores from both classifiers to measure and compare their estimated performance based on the OOB samples.\n",
        "\n",
        "#### Question:\n",
        "After training Random Forest models on both original and SMOTE-resampled datasets, how do the OOB scores compare, and what does this indicate about the impact of class balancing on model performance?\n",
        "\n",
        "#### Options:\n",
        "A) OOB score is higher for the original dataset, indicating better generalization without resampling.\n",
        "\n",
        "B) OOB score is higher for the SMOTE-resampled dataset, suggesting improved model performance due to better class balance.\n",
        "\n",
        "C) OOB scores are similar, indicating no significant impact from SMOTE resampling on model validation.\n",
        "\n",
        "D) OOB score is lower for the SMOTE-resampled dataset, which may indicate overfitting due to class oversampling.\n",
        "\n",
        "\n",
        "\n"
      ],
      "metadata": {
        "id": "cgE3IKCYjxvy"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "from sklearn.ensemble import RandomForestClassifier\n",
        "\n",
        "# TODO: Split the original data into training and testing sets\n",
        "X_train_orig, X_test_orig, y_train_orig, y_test_orig = train_test_split(X, y, test_size=0.33, random_state=42)\n",
        "\n",
        "# TODO: Split the resampled data into training and testing sets\n",
        "X_train_res, X_test_res, y_train_res, y_test_res = train_test_split(X_resampled, y_resampled, test_size=0.33, random_state=42)\n",
        "\n",
        "# TODO: Initialize a Random Forest classifier for the original data\n",
        "rf_classifier_orig = RandomForestClassifier(oob_score=True, random_state=42) # Hint: Set `oob score` to True\n",
        "\n",
        "# TODO: Train the classifier on the original data\n",
        "rf_classifier_orig.fit(X_train_orig, y_train_orig)\n",
        "\n",
        "# TODO: Retrieve the OOB score for the original data\n",
        "oob_score_orig = rf_classifier_orig.oob_score_ # Hint: Use documentation to access oob score.\n",
        "print(f\"The OOB score for original data is: {oob_score_orig:.4f}\")\n",
        "\n",
        "# Initialize a Random Forest classifier for the resampled data\n",
        "rf_classifier_res = RandomForestClassifier(oob_score=True, random_state=42) # Hint: Set `oob score` to True\n",
        "\n",
        "# TODO: Train the classifier on the resampled data\n",
        "rf_classifier_res.fit(X_test_res, y_test_res)\n",
        "\n",
        "# TODO: Retrieve the OOB score for the resampled data\n",
        "oob_score_res = rf_classifier_res.oob_score_\n",
        "print(f\"The OOB score for resampled data is: {oob_score_res:.4f}\")\n"
      ],
      "metadata": {
        "id": "QyaXur2egRuS",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "36ee32ce-a108-4f3e-80c8-b34b9f859213"
      },
      "execution_count": 13,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "The OOB score for original data is: 0.5553\n",
            "The OOB score for resampled data is: 0.7302\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "\n",
        "\n",
        "---\n",
        "\n",
        "## Most Accurate Class\n",
        "\n",
        "#### Context:\n",
        "Random Forest (RF) is a powerful ensemble classifier known for its accuracy and robustness. Understanding which class is most accurately predicted by the RF model can provide insights into the model's strengths and potential biases.\n",
        "\n",
        "#### Task:\n",
        "Use the previously trained rf_classifier, then analyze its predictions to determine which class is predicted most accurately based on recall from the classification report.\n",
        "\n",
        "#### Instructions:\n",
        "1. **Random Forest:** Use the previously trained rf_classifier on SMOTE data.\n",
        "2. **Predict Class Labels:** Use the trained classifier to predict labels for the test set.\n",
        "3. **Generate Classification Report:** Use `classification_report` from `sklearn.metrics` to generate detailed performance metrics for each class.\n",
        "4. **Analyze Recall Values:** Recall for each class indicates the proportion of actual positives that were correctly identified. Identify which class has the highest recall, signifying it as the most accurately predicted class.\n",
        "\n",
        "#### Question:\n",
        "Based on the Random Forest model's predictions, which class is the most accurately predicted in terms of recall?\n",
        "\n",
        "#### Options:\n",
        "A) Pastry\n",
        "\n",
        "B) Z_scratch\n",
        "\n",
        "C) K_scratch\n",
        "\n",
        "D) Stains\n"
      ],
      "metadata": {
        "id": "qZH3gQnH_B6_"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "from sklearn.metrics import classification_report, confusion_matrix\n",
        "\n",
        "# TODO: Predict the classes for the test set for the resampled data\n",
        "y_pred = rf_classifier_res.predict(X_test)\n",
        "\n",
        "# Generate the classification report\n",
        "class_report = classification_report(y_test, y_pred, output_dict=True)\n",
        "\n",
        "# Print the classification report for human reading\n",
        "print(class_report)\n",
        "\n",
        "# Find the class with the highest recall (indicating the most correctly classified)\n",
        "most_correctly_classified = max(class_report.keys(), key=lambda x: class_report[x]['recall'] if x != 'accuracy' and isinstance(class_report[x], dict) else 0)\n",
        "most_correctly_classified_score = class_report[most_correctly_classified]['recall']\n",
        "\n",
        "print(f\"The most correctly classified class is: {most_correctly_classified} with a recall of: {most_correctly_classified_score:.4f}\")"
      ],
      "metadata": {
        "id": "lZOLTDYb8QLD",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "38caca45-1100-46af-9d1d-60884139c06c"
      },
      "execution_count": 14,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "{'Bumps': {'precision': 0.7172958735733099, 'recall': 0.7012875536480687, 'f1-score': 0.709201388888889, 'support': 1165.0}, 'Dirtiness': {'precision': 0.38333333333333336, 'recall': 0.8598130841121495, 'f1-score': 0.5302593659942363, 'support': 107.0}, 'K_Scatch': {'precision': 0.9224719101123595, 'recall': 0.9636150234741784, 'f1-score': 0.9425947187141216, 'support': 852.0}, 'Other_Faults': {'precision': 0.745125348189415, 'recall': 0.6403351286654698, 'f1-score': 0.6887672996459607, 'support': 1671.0}, 'Pastry': {'precision': 0.6809269162210339, 'recall': 0.631404958677686, 'f1-score': 0.6552315608919382, 'support': 605.0}, 'Stains': {'precision': 0.6702127659574468, 'recall': 0.9692307692307692, 'f1-score': 0.7924528301886793, 'support': 130.0}, 'Z_Scratch': {'precision': 0.6524216524216524, 'recall': 0.8327272727272728, 'f1-score': 0.731629392971246, 'support': 275.0}, 'accuracy': 0.736108220603538, 'macro avg': {'precision': 0.6816839714012215, 'recall': 0.7997733986479421, 'f1-score': 0.7214480796135815, 'support': 4805.0}, 'weighted avg': {'precision': 0.7463519837510849, 'recall': 0.736108220603538, 'f1-score': 0.7362352286370477, 'support': 4805.0}}\n",
            "The most correctly classified class is: Stains with a recall of: 0.9692\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "\n",
        "\n",
        "---\n",
        "\n",
        "## Know your RF\n",
        "\n",
        "#### Context:\n",
        "Random Forest is an ensemble method that builds multiple decision trees and combines their predictions to improve accuracy and control overfitting. The placement of features within these trees, especially influential ones like 'Outside_X_index', can significantly affect the model's performance and interpretation.\n",
        "\n",
        "#### Task:\n",
        "Evaluate the statement regarding the placement of a strong predictor within the base learners of a Random Forest. The statement suggests that if 'Outside_X_index' is a very strong predictor, it will be the root node in all the base learners (Decision Trees) of the Random Forest.\n",
        "\n",
        "#### Question:\n",
        "Is the statement \"If 'Outside_X_index' is a very strong predictor of the output variable, then all of the base learners (DT) in a Random Forest will have this feature as the root node\" true or false?\n",
        "\n",
        "#### Options:\n",
        "A) True\n",
        "\n",
        "B) False\n"
      ],
      "metadata": {
        "id": "RWKVEVIgBNED"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "\n",
        "---\n",
        "\n",
        "## Optimizing RandomForest\n",
        "\n",
        "#### Context:\n",
        "Randomized Search is an efficient method for hyperparameter tuning that explores a given parameter space to find the optimal settings for a machine learning model. This approach helps enhance model performance by systematically testing different combinations of parameters.\n",
        "\n",
        "#### Task:\n",
        "Implement a Randomized Search CV to optimize a RandomForest Classifier using a specific set of hyperparameters. The goal is to identify the best combination of parameters that leads to the highest cross-validation score.\n",
        "\n",
        "#### Instructions:\n",
        "1. **Set Up Parameter Grid:** Define a parameter grid that includes potential values for 'n estimators', 'max depth', 'min samples split', and 'min samples_leaf'.\n",
        "2. **Initialize RandomForest:** Create a RandomForest classifier ensuring to set `random_state=42` for reproducibility.\n",
        "3. **Configure RandomizedSearchCV:** Set up RandomizedSearchCV with the RandomForest classifier, the defined parameter grid, and specify `n_iter=30` and `cv=2` for 30 iterations with 2-fold cross-validation. Also set `random_state=42`.\n",
        "4. **Execute the Search:** Fit the RandomizedSearchCV on your dataset to find the best parameter settings.\n",
        "5. **Report Results:** Output the best parameters and the best cross-validation score obtained.\n",
        "\n",
        "#### Parameters for Search:\n",
        "```python\n",
        "params = {\n",
        "    'n_estimators': [50, 100, 200],\n",
        "    'max_depth': [None, 10, 20, 30],\n",
        "    'min_samples_split': [2, 5, 10],\n",
        "    'min_samples_leaf': [1, 2, 4]\n",
        "}\n",
        "```\n",
        "\n",
        "#### Question:\n",
        "After performing Randomized Search with the specified parameters on a RandomForest model, what will be the best parameters?\n",
        "\n",
        "#### Options:\n",
        "A) {'n_estimators': 100, 'min_samples_split': 2, 'min_samples_leaf': 1, 'max_depth': 10}\n",
        "\n",
        "B) {'n_estimators': 50, 'min_samples_split': 2, 'min_samples_leaf': 1, 'max_depth': None}\n",
        "\n",
        "C) {'n_estimators': 200, 'min_samples_split': 2, 'min_samples_leaf': 1, 'max_depth': 10}\n",
        "\n",
        "D) {'n_estimators': 200, 'min_samples_split': 2, 'min_samples_leaf': 1, 'max_depth': 30}\n",
        "\n",
        "#### Note:\n",
        "The parameter grid provided here is intentionally limited due to time constraints and is mainly for practice purposes. To potentially achieve better results, consider expanding the range of parameters and allowing more time for tuning. The current setup is designed to provide a basic understanding of how parameter tuning affects model performance but does not guarantee optimal results."
      ],
      "metadata": {
        "id": "0uD_pbpnBf3E"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "from sklearn.model_selection import RandomizedSearchCV\n",
        "\n",
        "# Define a more extensive parameter grid for Random Search\n",
        "params = {\n",
        "    'n_estimators': [50, 100, 200],\n",
        "    'max_depth': [None, 10, 20, 30],\n",
        "    'min_samples_split': [2, 5, 10],\n",
        "    'min_samples_leaf': [1, 2, 4],\n",
        "}\n",
        "\n",
        "# Initialize the RandomForest classifier\n",
        "rf = RandomForestClassifier(random_state=42)\n",
        "\n",
        "# TODO: Setup RandomizedSearchCV with randomforest model,  30 iterations and 2-fold cross-validation\n",
        "random_search = RandomizedSearchCV(estimator=rf, param_distributions=params, n_iter=30, cv=2, verbose=1, random_state=42)\n",
        "\n",
        "# TODO: Perform the Random Search\n",
        "random_search.fit(X_train_res, y_train_res)\n",
        "\n",
        "# Print the best parameters and the best score\n",
        "print(\"Best parameters found: \", random_search.best_params_)\n",
        "print(\"Best cross-validation score achieved: \", random_search.best_score_)"
      ],
      "metadata": {
        "id": "EbAvo1kn0Fas",
        "outputId": "2bc25cd9-f84f-48ac-d460-26ad7a129bd9",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "execution_count": 20,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Fitting 2 folds for each of 30 candidates, totalling 60 fits\n",
            "Best parameters found:  {'n_estimators': 200, 'min_samples_split': 2, 'min_samples_leaf': 1, 'max_depth': 30}\n",
            "Best cross-validation score achieved:  0.7439684402712572\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "\n",
        "\n",
        "---\n",
        "\n",
        "## RF for Feature Importance\n",
        "\n",
        "#### Context:\n",
        "Random Forest (RF) classifiers not only provide robust predictive capabilities but also offer insights into feature importance within the dataset. This ability to evaluate feature significance is crucial for understanding the factors that most strongly influence the model's predictions.\n",
        "\n",
        "#### Task:\n",
        "Confirm if a RandomForest can be used to derive feature importances and identify the most important feature according to the best-performing model obtained from RandomizedSearchCV.\n",
        "\n",
        "#### Instructions:\n",
        "1. **Retrieve Best Model:** Obtain the best RandomForest model from the RandomizedSearchCV results.\n",
        "2. **Extract Feature Importances:** From the best model, extract the feature importance scores.\n",
        "3. **Identify and Sort Features:** Get the names of the features used in the model and sort them by their importance scores in descending order.\n",
        "4. **Visualize Feature Importances:** Generate a plot to visually display the importance of each feature, emphasizing the feature with the highest score.\n",
        "5. **Determine the Most Important Feature:** Identify and report the feature with the highest importance score.\n",
        "\n",
        "#### Question:\n",
        "Can Feature Importance be effectively applied using a RandomForest classifier? If so, which is the most important feature according to the best RF model from your analysis?\n",
        "\n",
        "#### Options:\n",
        "A) Yes, RF can be used for Feature Importance, and the most important feature is 'Outside-Global-Index'.\n",
        "\n",
        "B) Yes, RF can be used for Feature Importance, and the most important feature is 'Length_of_Conveyer'.\n",
        "\n",
        "C) Yes, RF can be used for Feature Importance, and the most important feature is 'Luminosity-lndex'.\n",
        "\n",
        "D) No, RF cannot be used for Feature Importance.\n",
        "\n"
      ],
      "metadata": {
        "id": "NlAQwPF7gmTZ"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# TODO: Get the best RandomForest model from RandomizedSearchCV\n",
        "best_rf_model = random_search.best_estimator_\n",
        "\n",
        "# TODO: Get feature importances from the best model\n",
        "feature_importances = best_rf_model.feature_importances_\n",
        "# Get feature names\n",
        "feature_names = X_train_res.columns  # Adjust this if your feature names are stored differently\n",
        "# TODO: Sort the features by their importance (and keep track of their names)\n",
        "sorted_indices = np.argsort(feature_importances)[::-1] # Hint sort only the `indicies` of the feature according to their importances.\n",
        "sorted_importances = feature_importances[sorted_indices]\n",
        "sorted_feature_names = feature_names[sorted_indices]\n",
        "\n",
        "# Print the most important feature and its importance score\n",
        "most_important_feature_name = sorted_feature_names[0]\n",
        "print(f\"The most important feature is: {most_important_feature_name} with an importance score of: {sorted_importances[0]:.4f}\")\n",
        "\n",
        "# TODO: Plot all feature importances sorted\n",
        "plt.figure(figsize=(12, 8))\n",
        "plt.barh(range(len(sorted_importances)), sorted_importances, align='center', color='skyblue')\n",
        "plt.yticks(range(len(sorted_importances)), sorted_feature_names)\n",
        "plt.xlabel('Feature Importance')\n",
        "plt.ylabel('Feature Names')\n",
        "plt.title('Sorted Feature Importance from the Best RandomForest Model')\n",
        "plt.gca().invert_yaxis()  # Invert y-axis to have the highest importance on top\n",
        "plt.show()"
      ],
      "metadata": {
        "id": "cmeAJkgkFxrr",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 735
        },
        "outputId": "fd09ba9b-aba8-4ebb-e759-17f07b327ea0"
      },
      "execution_count": 25,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "The most important feature is: Length_of_Conveyer with an importance score of: 0.0750\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1200x800 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [],
      "metadata": {
        "id": "kt8U1Tp7gprX"
      },
      "execution_count": null,
      "outputs": []
    }
  ]
}